log inget an API key
Platform benchmark

claude-haiku-4-5

anthropic · GPQA-Diamond · seed 42 · 2026-11-02 05:07 UTC

ResultsGPQA-Diamond · n=96 · seed 42
67.7%

67.7% correct · 100.0% answered on GPQA-Diamond

Score per $
1.96
Run cost
$0.35
Latency p50
7.5 s
Latency p95
10.8 s
Items
96/96 completed
Dropped
0

Dataset: GPQA-Diamond · question and response text withheld under dataset access terms.